Agent skill
eureka
Capture technical breakthroughs and transform them into actionable, reusable documentation. Use this skill when the user has achieved a significant technical insight, solved a hard problem, discovered a non-obvious solution, or wants to document a breakthrough moment. Also trigger when the user mentions "eureka", "breakthrough", "document this insight", "capture this discovery", or wants to turn a technical win into reusable knowledge.
Install this agent skill to your Project
npx add-skill https://github.com/feiskyer/claude-code-settings/tree/main/skills/eureka
SKILL.md
Eureka
Capture technical breakthroughs and transform them into actionable, reusable knowledge assets while context is fresh.
Process
1. Capture the Breakthrough
When invoked, immediately gather the essential details:
- What was the problem? The specific challenge or blocker faced.
- What was the insight? The key realization that unlocked the solution.
- What changed? Concrete before/after differences (metrics, behavior, code).
- Is there a minimal working example? Request one if not provided.
Ask clarifying questions if any of these are unclear, but act fast — capture while context is fresh.
2. Create the Breakthrough File
Create a structured markdown file at breakthroughs/YYYY-MM-DD-[brief-name].md using today's date and a concise kebab-case name derived from the breakthrough.
Use this template:
# [Descriptive Title]
**Date**: YYYY-MM-DD
**Tags**: [relevant, searchable, tags]
## One-Line Summary
A single sentence capturing the core insight.
## The Problem
What was the challenge? What wasn't working? Include specific error messages, performance numbers, or behavioral descriptions.
## The Insight
The key realization or discovery. What made this non-obvious? Why did previous approaches fail?
## Implementation
Minimal working code demonstrating the solution:
\`\`\`[language]
# concrete code example
\`\`\`
## Impact
| Metric | Before | After |
|--------|--------|-------|
| [relevant metric] | [value] | [value] |
## Reusable Pattern
Abstract the specific solution into a general principle that can be applied elsewhere:
- **When you see**: [symptom or situation]
- **Consider**: [the general approach]
- **Because**: [why it works]
## Related Resources
- Links to relevant documentation, issues, or discussions
3. Update the Index
Update breakthroughs/INDEX.md with a new entry. If the index file does not exist, create it with this structure:
# Breakthrough Index
| Date | Title | Tags | Link |
|------|-------|------|------|
| YYYY-MM-DD | [Title] | [tags] | [link to file] |
Append the new entry to the table, keeping entries in reverse chronological order (newest first).
4. Extract Reusable Patterns
After documenting, help the user identify:
- Is this pattern applicable to other parts of the codebase?
- Could this become a lint rule, test, or automated check?
- Should this be shared with the team?
Key Principles
- Act fast: Capture insights while context is fresh — don't over-polish on first pass.
- Be specific: Include concrete metrics, error messages, and code. Vague breakthroughs are not reusable.
- Think reusable: Always extract the generalizable pattern from the specific solution.
- Stay searchable: Use consistent tags and clear titles so breakthroughs can be found later.
Your Task
Document the breakthrough based on what the user describes:
- If the description is clear, proceed directly to creating the breakthrough file
- If details are missing, ask focused clarifying questions first
- Create the file, update the index, and highlight the reusable pattern
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
kiro-skill
Interactive feature development workflow from idea to implementation. Creates requirements (EARS format), design documents, and task lists. Triggers: "kiro", ".kiro/specs/", "feature spec", "需求文档", "设计文档", "实现计划".
github-review-pr
Review GitHub pull requests with detailed, multi-perspective code analysis using parallel subagents. Use this skill whenever the user wants to review a PR, asks for code review on a pull request, mentions "review PR", "check this PR", "look at pull request", or references a PR number or GitHub PR URL. Also trigger when the user wants feedback on code changes, wants to approve or request changes on a PR, or asks to review someone's contribution.
autonomous-skill
Execute long-running, multi-session tasks autonomously using Claude Code headless mode or in-session hook-based loops. Supports structured task decomposition (for complex projects) and lightweight Ralph-style iteration (for TDD, bug fixing, refactoring). Use this skill whenever the user says "autonomous", "long-running task", "multi-session", "run this in the background", "keep working on this", "batch process", "iterate until done", "ralph loop", or wants any task that requires sustained, unattended execution.
deep-research
深度调研的多Agent编排工作流:把一个调研目标拆成可并行子目标,用 Claude Code 非交互模式(`claude -p`)运行子进程;联网与采集优先使用已安装的 skills,其次使用 MCP 工具;用脚本聚合子结果并分章精修,最终交付"成品报告文件路径 + 关键结论/建议摘要"。用于:系统性网页/资料调研、竞品/行业分析、批量链接/数据集分片检索、长文写作与证据整合,或用户提及"深度调研/Deep Research/Wide Research/多 Agent 并行调研/多进程调研"等场景。
translate
Translate English or Japanese tech articles and texts into natural, fluent Chinese. Use this skill when the user wants to translate text to Chinese, asks for Chinese translation, mentions "translate to Chinese", "翻译", provides English/Japanese tech content for translation, or wants any text converted into Chinese. Also trigger when the user pastes text and asks to translate it, or references a file to translate into Chinese.
spec-kit-skill
GitHub Spec-Kit integration for constitution-based spec-driven development. 7-phase workflow. Triggers: "spec-kit", "speckit", "constitution", "specify", ".specify/", "规格驱动开发", "需求规格".
Didn't find tool you were looking for?